Project Summary: The United States is experiencing an unprecedented rate of opioid-related overdose death following a dramatic rise in opioid painkiller use over the past 20 years. In response, the CDC, several states, and professional organizations have promoted prescriber guidelines for managing pain with non-opioid alternatives (i.e., anti-inflammatories, anti-depressants). Additional emphasis has focused on reducing the impact of concomitant use of sedating medications (i.e. benzodiazepines) and other controlled substances (i.e. hypnotics, muscle relaxants) that synergistically increase risk of fatal overdose, especially in medically compromised individuals. Recent reports suggest that states with medical marijuana programs have lower rates of opioid overdose death. Possible mechanisms for lower rates of overdose include reductions in opioid use and/or the use of sedating medications due to symptomatic relief of comorbid anxiety or insomnia. However, little is known about the relationship between medical marijuana program participation among patients with pain conditions and changes in use patterns of controlled substance use. Even less is known about the differential impact of specific cannabinoid products medically dispensed (such as high tetrahydrocannabinol [THC] versus high cannabidiol [CBD] products). The New York State medical marijuana program (implemented January 2016) is one of the most highly regulated and medicalized programs in operation in the United States. State-licensed dispensaries must be staffed with licensed pharmacists and demonstrate that they dispense products with consistent cannabinoid ratios (including high-THC, high-CBD, and 1:1 THC:CBD products). New York State is therefore an ideal setting for a rigorous analysis of medical cannabis program participation and the impact of cannabinoid products with varying THC:CBD ratios on controlled substance use over time among chronic pain patients. We propose to partner with the largest medical marijuana provider in New York (> 10,000 participants enrolled as of October 2017). Participant activity is tracked through an electronic health record with de-identified patient characteristics at baseline (demographics, qualifying condition and symptoms, concomitant medications) that is linked to data containing 12-month controlled substance prescription history through the state?s prescription drug monitoring program (PDMP). The analysis of these de-identified, compiled records is a highly innovative and cost-effective method for better understanding the potential mechanisms of how participation in medical marijuana programs influences controlled substance use and overdose risk among patients with pain conditions receiving opioids and sedative-hypnotics and can inform subsequent prospective study designs.